Techniques for Binarizing Images to Determine Jetting Quality in Additive Manufacturing

1. A method, comprising: obtaining an image of a jet of material being jetted from a nozzle of a 3D printer during a printing process; binarizing the image to distinguish background features from foreground features contained in the image, the foreground features depicting the jet of material; and determining, by a processing device, a jetting quality of the jet of material based on the binarized image.

2. The method of claim 1, wherein binarizing the image comprises: computing, by a processing device, a gradient that describes a change in a number of background or foreground pixels in the image as a function of an intensity threshold value; determining a final intensity threshold value based on the gradient; and binarizing the image using the final intensity threshold value.

3. The method of claim 2, wherein computing the gradient comprises iteratively incrementing a value of the intensity threshold value, and at each increment: binarizing the image using the intensity threshold value; and computing a number of foreground pixels in the binarized image, wherein the gradient describes a change in the number of foreground pixels as a function of change in the intensity threshold value.

4. The method of claim 2, wherein determining the final intensity threshold value comprises identifying a minimum of the gradient, and identifying a start of a plateau in the gradient after the minimum.

5. The method of claim 1, wherein binarizing the image comprises: dividing the image into a jetting-free domain that is expected to not include foreground objects and a jetting path domain that is expected to include foreground objects; computing statistical data for the jetting-free domain comprising a mean pixel-level intensity and a standard deviation pixel-level intensity; and binarizing the jetting path domain using the statistical data in the jetting-free domain.

6. The method of claim 5, wherein binarizing the jetting path domain comprises comparing each pixel intensity against the statistical data obtained from the jetting-free domain, wherein a particular pixel is determined to be part of a background object if the pixel intensity of the particular pixel falls within a prescribed confidence interval of intensity values defined from the statistical data of the jetting-free domain.

7. The method of claim 5, wherein the jetting-free domain comprises a left domain to a left of the jetting path domain and a right domain to a right of the jetting path domain.

8. The method of claim 5, wherein binarizing the jetting path domain comprises generating a z-score for a pixel using the statistical data and comparing the z-score to a threshold to determine whether the pixel is a foreground pixel or a background pixel.

9. The method of claim 5, wherein computing the statistical data for the image and binarizing the image using the statistical data is performed individually for each row of pixels in the image.

10. The method of claim 1, wherein binarizing the image comprises: generating a matrix of pixel intensities for the image; and fitting, by a processing device, a polynomial function to the matrix of pixel intensities and subtracting the polynomial function from the image to generate a flattened image.

11. A non-transitory computer-readable storage medium having instructions stored thereon that, when executed by a processing device, cause the processing device to: obtain an image of a jet of material being jetted from a nozzle of a 3D printer during a printing process; binarize the image to distinguish background features from foreground features contained in the image, the foreground features depicting the jet of material; and determine a jetting quality of the jet of material based on the binarized image.

12. The non-transitory computer-readable storage medium of claim 11, wherein the instructions that cause the processing device to binarize the image cause the processing device to: compute a gradient that describes a change in a number of background or foreground pixels in the image as a function of an intensity threshold value; determine a final intensity threshold value based on the gradient; and binarize the image using the final intensity threshold value.

13. The non-transitory computer-readable storage medium of claim 12, wherein to compute the gradient comprises to iteratively increment a value of the intensity threshold, and at each increment: binarize the image using the intensity threshold value; and computing a number of foreground pixels in the binarized image, wherein the gradient describes a change in the number of foreground pixels as a function of change in the intensity threshold value.

14. The non-transitory computer-readable storage medium of claim 12, wherein to determine the final intensity threshold value comprises to identify a minimum of the gradient, and identify a start of a plateau in the gradient after the minimum.

15. The non-transitory computer-readable storage medium of claim 11, wherein the instructions that cause the processing device to binarize the image cause the processing device to: divide the image into a jetting-free domain that is expected to not include foreground objects and a jetting path domain that is expected to include foreground objects; compute statistical data for the jetting-free domain comprising a mean pixel-level intensity and a standard deviation pixel-level intensity; and binarize the jetting path domain using the statistical data in the jetting-free domain.

16. The non-transitory computer-readable storage medium of claim 15, wherein to binarize the jetting path domain comprises to compare each pixel intensity against the statistical data obtained from the jetting-free domain, wherein a particular pixel is determined to be part of a background object if the pixel intensity of the particular pixel falls within a prescribed confidence interval of intensity values defined from the statistical data of the jetting-free domain.

17. The non-transitory computer-readable storage medium of claim 15, wherein the jetting-free domain comprises a left domain to a left of the jetting path domain and a right domain to a right of the jetting path domain.

18. The non-transitory computer-readable storage medium of claim 15, wherein to binarize the jetting path domain comprises to generate a z-score for a pixel using the statistical data and compare the z-score to a threshold to determine whether the pixel is a foreground pixel or a background pixel.

19. The non-transitory computer-readable storage medium of claim 15, wherein to compute the statistical data for the image and binarize the image using the statistical data is performed individually for each row of pixels in the image.

20. The non-transitory computer-readable storage medium of claim 11, wherein the instructions that cause the processing device to binarize the image cause the processing device to: generate a matrix of pixel intensities for the image; and fit a polynomial function to the matrix of pixel intensities and subtracting the polynomial function from the image to generate a flattened image.